Can a legendary engineer create immersive audio mixes without giving up his beloved SSL console? Of course he can!
When stereo replaced mono, it took time before mixing conventions crystallised. Half a century on, we find ourselves in a similar situation with respect to immersive audio. There’s a strong demand for immersive mixes from streaming services, labels and artists, but there’s still relatively little agreement about how that demand should be met. Should music mixers employ the centre speaker and, if so, to what end? How should sources be divided between objects and beds? Should an immersive mix be a separate process, working from stems, or should it be done concurrently with the stereo mix? And should the surround and height speakers carry fundamental musical information, or should they serve a secondary role replaying ambiences and effects?
Ask 10 engineers working in Atmos, and you’ll get 10 different answers to all of these questions. Ask Bob Clearmountain, and you’ll get a particularly interesting and unique perspective.
Bob Clearmountain is, of course, one of the most successful engineers in the history of recorded music. He practically invented the role of the specialist mix engineer, more than 40 years ago, and since then his name has appeared on an extraordinary number of hit singles and albums. And, like many mix specialists, he’s embraced immersive audio. However, his methods and his philosophy are entirely his own.
One of the pioneering figures from the early days of stereo was Bill Putnam Sr, founder of Universal Audio. Putnam saw the potential of stereo well before most of his clients did so, and adopted a policy of mixing everything in the new format even when labels hadn’t requested it — a piece of foresight which paid off handsomely in subsequent years. And for the last two decades, Bob Clearmountain has quietly been taking a similar approach with respect to surround sound.
“When 5.1 was a thing for about a minute, 20 years ago or more, I didn’t want to know, because I thought ‘Jesus, people have trouble getting two speakers in phase, much less five or six!’ But my wife, Betty from Apogee Electronics, said, ‘You should be doing this, because this is the future.’ And so I woke up in the middle of the night one night and figured out how I could do that. I’m mixing on an analogue desk, and I realised how we could modify the desk and do it simultaneously while I’m doing the stereo. It really only added about maybe 10 minutes, tops, to the time it took me to do a stereo mix, because I was doing it pretty much simultaneously.”
Unlike stereo, 5.1 surround came and went without achieving mass‑market popularity, but Bob Clearmountain nevertheless kept on mixing everything in both formats. “It was so easy to do, and so much fun, and I really enjoyed the way it sounded, so I just kept doing it. I’ve got a whole server full of 5.1 mixes — pretty much they go along with every stereo mix I’ve done.”
And for someone who was already working in 5.1 every day, the initial step to Atmos was relatively small. “I had mixed the first three albums from a group called the Band who were really popular back in the ’60s for each of these albums’ 50th anniversary in 5.1. BMG and Robbie Robertson from the Band asked me to do that, and so those actually did come out in 5.1 recently. But for the fourth one, which is called Cahoots, they called up and they said, ‘OK, for this one, could you do an Atmos mix?’ And that first Band album came out really well, I think. I was still learning what to do with it, but it wasn’t that much different than 5.1 because I’d mixed hundreds of songs in 5.1 and so it’s just more places to put stuff!”
Bob Clearmountain has a relaxed attitude towards the adaptation of his studio for surround mixing. “In the early days I was involved in a paper about the proper way to set up a 5.1 system for mixing along with a bunch of other engineers like George Massenburg, and I remember there was a big discussion about ‘Where do you put the rear speakers?’ and ‘What’s the angle in relation to the centre?’ And I said, ‘Man, I put them where nobody will bump into them.’ I’m just using basically large bookshelf Dynaudio BM15As and then they’re on speaker stands in the only place that I can fit them and it works perfectly, you know? It’s been fine. A lot of it’s just practicality, you know?”
He adopted the same pragmatic approach to Atmos monitoring, with its additional height speakers. “Because the studio’s in the basement of my house, we had just redone the floors in the living room, which is right above where I’m sitting. While we did that, I put these pipes in, thinking, ‘If I ever have to put speakers in the ceiling, now I have a way of getting wires to them instead of them hanging off the walls,’ you know? And luckily I did that. So I added a couple more speakers to my 5.1 system so I had a 7.1 system. I found these little Sony speakers on the internet and stuck them on the ceiling. It’s a low ceiling so I didn’t want any really large speakers hanging off the ceiling. I didn’t want tall people to hit their head!”
Most engineers who work in Atmos do so entirely ‘in the box’, using only the software mixer in their DAW and plug‑in effects. If any hardware is employed, it’s usually a control surface such as an Avid S‑series or Yamaha Nuage. Bob Clearmountain’s approach stands in complete contrast: he still mixes on an SSL G‑series analogue console, and although he uses plug‑ins, he treats these just like hardware effects, patching the computer into the console and sending from its auxes. The key to making this work is an ingenious but simple modification which allows him to repurpose the multichannel groups as additional outputs feeding the various surround speakers. Each channel can be assigned to any speaker pair, with the ‘small fader’ serving as a pan control between them.
“Most inline consoles nowadays have a main fader and a small fader. They call it a monitor fader, but you can do all kinds of things with it. It can be a post‑fader send to the multichannel groups, so I’ve put them all at unity, and they’re all post large fader to the groups, and I’ve chosen a group of those multichannel groups as sends to the Atmos channels. And I set it up in a way that I can have a static pan between any pair of speakers. I can pan something between the centre and the left front, or the left front and the left side, or the left side and the left rear or something like that. It’s like any pair of speakers I can pan because the way that works, and not just the SSL, but Neves and most of them, is you can pan between odd and even buses and so if you arrange them in the right way, you have a lot of versatility.
“Then I’ll have two extra buses. This console has what they call a Bluestone mod, so instead of 32 buses, it’s really only got 24, because the last eight are now auxes. The first three are also effects sends, and they do different things. They’re subgroups for effect sends and various other things that I do, so I only had two extra after all that, and those are my sort of ‘floating objects’. Actually I could have four, but mainly I have two or three and I think there’s only one time I’ve used both of those so far, after mixing like 10 albums in Atmos, so it just does doesn’t come up very much for me.”
The point of this setup is that it makes use only of console features that aren’t employed at all in Clearmountain’s stereo mixes. So, at its most basic, generating a 5.1 or Atmos mix from a stereo mix is a matter of simply pressing buttons to assign desk channels to different multichannel groups. This doesn’t affect what is heard at the main mix bus, so crucially doesn’t impact or constrain his stereo mix process in any way. What it does mean is that Clearmountain’s Atmos mixes are predominantly channel‑based rather than object‑based. Some would see this as a limitation, but it’s not a problem for Bob; in fact, you get the sense that he’d often be happy not to use objects at all, if it wasn’t for the fact that the Atmos bed channels don’t differentiate between the front and rear height speakers. “In the Atmos bed, the overheads are just stereo. There’s no quad. There aren’t four channels, so my height speakers are actually four basically stationary objects.”
Clearmountain’s willingness to rely on beds isn’t so much a matter of mixing philosophy as one of genre. Classic, guitar‑based rock and pop music makes up the bulk of his workload, and he rarely sees a need to have lots of different sources flying around the listener’s head...
Clearmountain’s willingness to rely on beds isn’t so much a matter of mixing philosophy as one of genre. Classic, guitar‑based rock and pop music makes up the bulk of his workload, and he rarely sees a need to have lots of different sources flying around the listener’s head. He’s quite willing to acknowledge that other styles demand a different approach. “A friend of mine came over, he had done a mix in the box. It was a female singer, but the rest of it was basically EDM, and they had stuff flying all over the place, you know, and things going mono and then surround and back and forth and spinning and it sounded amazing. I mean, I was really impressed. It was totally the opposite of anything that I would ever do, but I loved it. I just thought this is really cool and it’s nice that there’s so many different ways of approaching this, you know?”
One of the other aspects of Atmos mixing on which consensus has yet to be reached concerns the use of the centre speaker. If you want a source such as a lead vocal to appear right in front of the listener, routing it only to the centre speaker will achieve that, but the ‘phantom centre’ effect works in Atmos too, and many people advocate using only the left and right speakers. This is a point on which Clearmountain has strong opinions.
Bob Clearmountain: "One of my favourite things is the centre speaker, because the nice thing is when you anchor stuff to speakers, especially the centre, you can walk around the room and it doesn’t move."
“That’s just silly. I think. One of my favourite things is the centre speaker, because the nice thing is when you anchor stuff to speakers, especially the centre, you can walk around the room and it doesn’t move. If it’s just phantom, you walk over to the right and the phantom centre follows you to the right, just like it does in stereo — which is one of the drawbacks of stereo. I like actually walking around the room, I’ll stand over here on the side between the right side and right rear and the picture still stays the same. I mean, the balances are different, so I’ll be hearing more of whatever’s coming out of those speakers, but everything’s still in the same place, right? The vocal’s still coming from the centre, and I love that.
“I do a little divergence, where I put a bit of it down about 15 or 20 dB in the left and right so it’s not totally in the centre, but basically you hardly even notice that, it just makes it a bit fuller sounding that way — but I’m a big fan of the centre. One problem is that some labels don’t like you to do that because then people can strip out the vocal by itself and then do their own little tricks with it, but I don’t concern myself with things like that, that’s somebody else’s problem!”
Keeping vocals and other lead instruments mainly in the centre speaker allows Clearmountain to place vocal effects in the left and right speakers — and frequently the surrounds as well. “To me, reverb is not coming from where the voices are. I really enjoy spreading the effects around. Most of my effects are stereo or quad or more than that. With Atmos, I usually have a quad ambience or a reverb in the ceiling, and that kind of makes the whole picture a little bit bigger. You’re not necessarily that aware of it. It’s just that if you turn it off, you realise: ‘Oh, what happened to that? The room just got smaller!’ I mean, I will fly things around once in a while. If it’s some crazy, psychedelic guitar solo that might take a little trip around the room, things like that. I did this Joe Bonamassa record where he does this backwards kind of ’60s psychedelic guitar solo, and that spins around, but there’s delays coming off it and so you know it’s moving, you still get this ambient thing happening and it actually makes it more interesting, I think, than just having a single sound swimming around the room. It makes it sound more realistic, like it’s actually in the room.
“I mix a lot of live concert videos and live albums. Maybe half the stuff I normally do is live performances. And that’s usually just the band on the stage, which are in the front three speakers, mostly, and the rest of it’s audience and ambience. And I’ll add to that ambience: I’ll put little delays up in the top rear speakers or in the back, and different reverbs and things like that. Plus the fact that you have the audience mics. Hopefully you have more than two audience mics. Some of the older stuff… I just did the Stones album from the El Mocambo in Toronto. That was recorded in ’77, of course it was just stereo audience track back then, so those I put in the sides and then delayed them to the rears a little bit. At first I thought, ‘Man, is that going to work? Is it just going to sound like a delay?’ And it actually doesn’t, it works pretty well.
“Once in a while, sometimes on a rock song where you want an old‑fashioned ’50s or ’60s‑type slap, I’ll just stick that in the centre, or maybe I’ll use a spring reverb to get a real kind of lonely sort of a sound on a vocal where I’ll just tie that in with the vocal right in the centre.”
Although Atmos is being pushed hard by Apple Music and other streaming services, it’s only a tiny minority of listeners who are consuming it on a 7.1.4 loudspeaker set up. The vast majority of Atmos listening takes place on headphones, more so even than in stereo. Bob Clearmountain seems resigned to the fact that this will never offer quite the same listening experience as a good speaker‑based rig. “Because of the way the Apogee Symphony MkII works, it’s really easy to A/B between my stereo mix and the Atmos mix or the binaural or whatever, and so I’ll just try to make sure that it sounds as good as possible compared to the stereo mix. It generally does a funny thing with binaural, because it’s trying to make things sound like they’re coming from behind you or above you or whatever, and so it tends to mess with the frequency response a bit. It has all these binaural settings that you can set to off, near, mid and far for each of the channels, and the farther you set things away, they sound kind of dull. It actually adds a little bit of ambience, which is really disconcerting to me, because I’m kind of a freak about my effects and ambience, and I don’t really like the recording medium making decisions like that!”
Unfortunately, Bob’s own experiences suggest that it will be a long time before most home listeners get to experience speaker‑based Atmos in all its glory. “I wish home audio manufacturers would come up with a way for people to set Atmos up in their living rooms that’s easy and that’s wireless. I have the Sonos system; that’s a sound bar which is basically three speakers, and then it has a speaker that points up at the ceiling for the height, and a sub and then these two little point‑one speakers. The system sounds really good. It’s basically a 5.1 system, though, and I tried to play something off Apple Music the other day in surround, and you have to plug in an Apple TV and you have to have a TV with an ARC output — and I couldn’t get it to work. It would only play stereo!
“If I can’t do it — and I’m a recording engineer, right? — how does somebody at home figure out how to do this? I wish people could hear it the way we’re mixing it, and very few people will ever hear it that way unless something changes, unless somebody gets it together and makes it work easier. You shouldn’t have to be a graduate of MIT to be able to play something back!”
An extended version of this interview is available as a podcast in the SOS Recording & Mixing channel: www.soundonsound.com/sos-podcasts
One of the features that endeared the SSL console to mix engineers back in the late ’70s was its stereo bus compressor. And one of the things that many engineers find hard to adjust to in Atmos mixing is that there is no stereo bus as such. Bob Clearmountain still leans heavily on the SSL compressor in his stereo mixes — but he’s also found a way to use it in 5.1 and Atmos.
“The SSL happens to have eight extra patchable VCAs that are the same VCAs that are in the compressor, so I thought, well, I’m sure there’s a way I could hook this up so I could use them for my surround mix. So I went to my friend Lucas van der Mee over at Apogee and he said, ‘Yeah, sure, we could just stick a little op‑amp in there and we can slave the patch VCAs off of the bus compressor.’ So, as long as I have a stereo mix going into the main stereo bus compressor, then the surround compressor’s just following along. And it worked quite well.
“We just took the DC voltage out of the main bus compressor and then boosted it with an op‑amp and it’s just running all the rest of the stuff, and it sounds really good. I still have that same exciting compression sound that the stereo bus compressor adds. I enjoy fast‑release compressors — that’s why I’m not a big fan of the Fairchild, which is the thing that everybody bows down to. To me, the thing’s kind of useless because it’s just too slow. I change the ratio occasionally; if it’s a ballad I’ll do less and then if it’s a rock song I’ll go between 2:1 and 4:1 and then adjust the threshold control, but the attack and release stay pretty much the same.
“Of course, then Atmos came along and I said, ‘Oh shit, now what am I going to do?’ And my clever assistant, Brandon Duncan, went on a used gear site and he actually found another Eurorack, which is the rack in the SSL that holds these stereo VCA cards. He found a complete rack with the cards. So I bought that and between Brandon and Lucas, they figured out a way to wire it all up so I have now basically a 16‑channel bus compressor. And not only that, but Lucas added three extra controls. If I’m mixing something that’s live with an audience, sometimes if I’m hitting that thing too hard, it sounds like the audience is kind of fluttering in and out because the rear channels are mostly audience — and so now I can reduce how much compression’s going to the sides, the rears or the overheads.”